Write a program in Python to find the index for NaN value in a given series

When working with data analysis in Python, you often encounter NaN (Not a Number) values in your datasets. Finding the indices where these NaN values occur is a common task. This guide shows multiple approaches to locate NaN indices in a Pandas Series.

Creating a Sample Series

Let's first create a sample series containing NaN values ?

import pandas as pd
import numpy as np

l = [1, 2, 3, np.nan, 4, np.nan]
data = pd.Series(l)
print(data)

0    1.0
1    2.0
2    3.0
3    NaN
4    4.0
5    NaN
dtype: float64

Method 1: Using Loop with np.isnan()

This method iterates through the series and checks each value using np.isnan() ?

import pandas as pd
import numpy as np

l = [1, 2, 3, np.nan, 4, np.nan]
data = pd.Series(l)

for i, j in data.items():
    if np.isnan(j):
        print("index is", i)

index is 3
index is 5

Method 2: Using isna() with Index

A more Pythonic approach using Pandas built−in methods ?

import pandas as pd
import numpy as np

l = [1, 2, 3, np.nan, 4, np.nan]
data = pd.Series(l)

nan_indices = data.index[data.isna()]
print("NaN indices:", list(nan_indices))

NaN indices: [3, 5]

Method 3: Using np.where()

Using NumPy's where() function to find NaN positions ?

import pandas as pd
import numpy as np

l = [1, 2, 3, np.nan, 4, np.nan]
data = pd.Series(l)

nan_indices = np.where(data.isna())[0]
print("NaN indices:", nan_indices)

NaN indices: [3 5]

Comparison

Conclusion

Use data.index[data.isna()] for most cases as it's readable and efficient. For large datasets, np.where() offers the best performance. The loop method is useful for learning but avoid it in production code.